• Replies 6 replies
  • Subscribers 960 subscribers
  • Views 299 views
  • Users 0 members are here
Related

Geo Area Reporting

Madeline Dummerth (Past Member)
$organization

I am curious about how other organizations report their geo area percentages. So say I want to have a pie chart that shows the breakdown of which areas our subscribers attend from. Do you use unique constituent ID or total ticket count? We are having a philosophical debate at my organization about this. 

Parents
  • $organization

    Great question, I think it has a lot to do with what you are trying to claim.

    Is your interest in which areas are most valuable to you organization financially, then total payed amount.

    or is your interest in mission then ticket counts might make the point about total impact

    But, constituent count is sort of a measure of the number of decision makers in an area who are choosing your organization.  Marketing might want to know this.  However this might lead to confusion.  One constituent might purchase 1000 tickets.  Or 500 constituents might each purchase two tickets each. As marketers your approach might be different.

    So I invite you to try to clarify what it is you are trying to claim or say about the data.

    Then to the question the type of chart.  It is my experience that pie charts are about a percentage of a whole by category.  There is sort of an assumption that each of the wedges are equally likely.  However in a geo area scenario that assumption might not be met.  Let’s say that one of your geo areas is a high density state very close to your organization  with 10 million residents and you got 1000 seats sold in that area.  And another one of your geo areas is a rural small town of 10 thousand people.  What would it mean if both had 1000 seats sold.  On a pie chart they would have exactly the same wedge size.  

    Recently, I have been playing with the scatter map chart in Tessitura Analytics.  I’ve been setting the geographic granularity  to zip code. (Short postal code) Because zip codes were created to support the number of customers a group of carriers could deliver in a day.  That tends to keep one from having a state pitted against a rural county. (This is not at all perfect, however it’s a starting point.  A GIS geographic information application might allow you to do a lot better)  

    I’m then making the color of the spots the average or median price paid for a ticket.  Depending on the distribution of prices. And the size of the dot the total count of tickets sold in that area.  

    This means that I can see dots scattered around a map that represents not too grossly different numbers of people per dot.  The color of the dots show the financial value of the area, and the size of the dot represents the amount of impact in the area.  

    Tech tip use the short postal code for this in Tessitura Analytics.  If you use the long postal code and have any zip +4 postal codes in your system.  In 15.0.4 the constituents with the zip+4 zip codes seem to be dropped.  

    Hope that helps.  We can discuss further at analytic Coffee! later today.

Reply
  • $organization

    Great question, I think it has a lot to do with what you are trying to claim.

    Is your interest in which areas are most valuable to you organization financially, then total payed amount.

    or is your interest in mission then ticket counts might make the point about total impact

    But, constituent count is sort of a measure of the number of decision makers in an area who are choosing your organization.  Marketing might want to know this.  However this might lead to confusion.  One constituent might purchase 1000 tickets.  Or 500 constituents might each purchase two tickets each. As marketers your approach might be different.

    So I invite you to try to clarify what it is you are trying to claim or say about the data.

    Then to the question the type of chart.  It is my experience that pie charts are about a percentage of a whole by category.  There is sort of an assumption that each of the wedges are equally likely.  However in a geo area scenario that assumption might not be met.  Let’s say that one of your geo areas is a high density state very close to your organization  with 10 million residents and you got 1000 seats sold in that area.  And another one of your geo areas is a rural small town of 10 thousand people.  What would it mean if both had 1000 seats sold.  On a pie chart they would have exactly the same wedge size.  

    Recently, I have been playing with the scatter map chart in Tessitura Analytics.  I’ve been setting the geographic granularity  to zip code. (Short postal code) Because zip codes were created to support the number of customers a group of carriers could deliver in a day.  That tends to keep one from having a state pitted against a rural county. (This is not at all perfect, however it’s a starting point.  A GIS geographic information application might allow you to do a lot better)  

    I’m then making the color of the spots the average or median price paid for a ticket.  Depending on the distribution of prices. And the size of the dot the total count of tickets sold in that area.  

    This means that I can see dots scattered around a map that represents not too grossly different numbers of people per dot.  The color of the dots show the financial value of the area, and the size of the dot represents the amount of impact in the area.  

    Tech tip use the short postal code for this in Tessitura Analytics.  If you use the long postal code and have any zip +4 postal codes in your system.  In 15.0.4 the constituents with the zip+4 zip codes seem to be dropped.  

    Hope that helps.  We can discuss further at analytic Coffee! later today.

Children